Platelet-activating factoracetylhydrolase and other novel risk and protective factors for cardiovascular disease in systemic lupus erythematosus.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 9, September 2004, pp 2869–2876 DOI 10.1002/art.20432 © 2004, American College of Rheumatology Platelet-Activating Factor–Acetylhydrolase and Other Novel Risk and Protective Factors for Cardiovascular Disease in Systemic Lupus Erythematosus Anna Cederholm,1 Elisabet Svenungsson,2 Dominique Stengel,3 Guo-Zhong Fei,1 A. Graham Pockley,4 Ewa Ninio,3 and Johan Frostegård1 Objective. There is an important inflammatory component to atherosclerosis and cardiovascular disease (CVD). It is therefore interesting that the risk of CVD is high in inflammatory diseases such as systemic lupus erythematosus (SLE). In this study, we investigated nontraditional risk factors for the development of CVD in patients with SLE. Methods. Twenty-six women (mean age 52 years) with SLE and a history of CVD were compared with 26 age-matched women with SLE and no clinical manifestations of CVD (SLE controls) and 26 age-matched healthy women (population controls). Serum levels of several novel nontraditional risk and protective factors were determined: heat-shock protein (HSP)–related factors (Hsp60, Hsp70, anti–human Hsp60, anti–human Hsp70, and anti–mycobacterial Hsp65), plateletactivating factor–acetylhydrolase (PAF-AH) activity, secretory phospholipase A2 GIIA (sPLA2), and anti– endothelial cell antibody (AECA). The intima-media thickness and the presence of plaques in the common carotid arteries were determined by B-mode ultrasound as a surrogate measure of atherosclerosis. Results. Levels of PAF-AH, but not HSP-related factors, AECA, or sPLA2, were significantly increased in SLE cases. Only PAF-AH discriminated between SLE cases and SLE controls (P ⴝ 0.005). PAF-AH was significantly associated with low-density lipoprotein (LDL) cholesterol and total cholesterol in the SLE cases (r ⴝ 0.50, P ⴝ 0.0093 and r ⴝ 0.54, P ⴝ 0.0045), but not in either control group. Conclusion. The increased levels of PAF-AH in SLE cases and the association between PAF-AH and LDL cholesterol adds support to the notion that PAF-AH may promote atherothrombosis in SLE. The role of HSPs in CVD is complex, since anti-Hsp65 appears to be associated with the presence of CVD, whereas Hsp70 might protect against it. In this crosssectional study, levels of HSP-related factors, AECA, and sPLA2 were not associated with CVD in SLE. Atherosclerosis, which together with thrombosis is a primary cause of cardiovascular disease (CVD), has many characteristics in common with inflammatory diseases. These include an abundant production of proinflammatory cytokines and the presence of immunocompetent cells in atherosclerotic lesions (1). A growing body of evidence indicates that patients with systemic lupus erythematosus (SLE) are at high risk of developing CVD, and this has been interpreted as being a late complication that has only become apparent with the better treatments available for this patient group and the consequential improvements in life expectancy (2,3). We have previously reported that although traditional risk factors such as dyslipidemia were typical of SLE-related CVD, other factors, including some antiphospholipid-related antibodies, lipid oxidation, and emerging inflammatory risk factors such as C-reactive protein, were also associated with SLE-related CVD (4). Studies of SLE-related CVD might therefore provide Supported by the King Gustaf V 80th Birthday Fund, the Swedish Society of Medicine, the Swedish Rheumatism Association, the Torsten and Ragnar Söderberg Foundation, the Swedish Science Fund, and the Swedish Heart-Lung Foundation. 1 Anna Cederholm, MD, Guo-Zhong Fei, MD, PhD, Johan Frostegård, MD, PhD: Karolinska University Hospital, Huddinge, Sweden; 2Elisabet Svenungsson, MD: Karolinska University Hospital, Solna, Sweden; 3Dominique Stengel, PhD, Ewa Ninio, PhD: INSERM U525/IFR14 Coeur Muscle Vaisseaux, and Université Pierre et Marie Curie/Faculté de Médecine Pitié-Salpêtrière, Paris, France; 4 A. Graham Pockley, PhD: University of Sheffield, Sheffield, UK. Address correspondence and reprint requests to Johan Frostegård, MD, PhD, Department of Medicine, Karolinska University Hospital, 141 86 Stockholm, Sweden. E-mail: email@example.com. Submitted for publication March 10, 2004; accepted April 29, 2004. 2869 2870 CEDERHOLM ET AL important insights into the role of immunologic and inflammatory factors in CVD, especially since laboratory animal models do not generally develop atherothrombotic disease/CVD that is typical of the condition in humans (5). We and other investigators have recently identified novel immunologic and inflammatory factors that may play important roles as putative risk factors and as causative agents in CVD and atherosclerosis. These include Hsp60, antibodies against mycobacterial Hsp65 (6–11), antibodies against endothelial cells (AECAs) (12), platelet-activating factor–acetylhydrolase (PAFAH) (13), and secretory phospholipase A2 (PLA2) (14,15). Interestingly, these novel factors are interrelated in several ways, and their roles in atherothrombosis may be complex and vary during different stages of disease development. Heat-shock proteins (HSPs), or stress proteins, are evolutionarily conserved molecules that are expressed on activated endothelium. These proteins are also immunogenic, and immune responses to mammalian and bacterial HSPs cross-react. In a previous study from our laboratory, AECA was found to be associated with anti-Hsp65 (12), and it has been reported that anti-HSP reacts with endothelium (16). Furthermore, oxidized LDL (ox-LDL), which the “oxidation hypothesis” suggests is a major factor in atherogenesis (17), induces the expression of HSPs (7). PAF-AH (also called lipoprotein-associated phospholipase A2 and LDL-PLA2) and secretory PLA2 are also important in the degradation of PAF and PAF-like lipids as well as in the generation of lysophosphatidylcholine (LPC). Interestingly, PAF-like lipids and LPC are both major proinflammatory components of ox-LDL that could induce and promote the inflammatory reaction in the vessel wall (15,18,19). AECAs cross-react with ox-LDL and LPC (20) and may be of major importance in vasculitis (21). In the present report, we provide clinical evidence that PAF-AH activity is increased in SLE patients with CVD, whereas levels of the other nonclassic risk factors we studied are not. The implications of these findings are discussed. MATERIALS AND METHODS Study group. The SLE cases consisted of 26 women with SLE who had 1 or more manifestations of CVD, defined as a history of myocardial infarction (n ⫽ 7), angina (n ⫽ 9), cerebral infarction (n ⫽ 15), or claudication (n ⫽ 4). The SLE controls consisted of 26 age-matched women with SLE and no clinical manifestations of CVD. The population controls con- sisted of 26 age-matched healthy women who were recruited randomly from the population registry. None of these subjects had arterial disease or SLE. Details of the recruitment and clinical characteristics of the 3 study groups have been reported previously (4). All patients fulfilled the American College of Rheumatology (ACR) 1982 revised criteria for the classification of SLE (22). Myocardial infarction was confirmed by electrocardiography and by an increase in the creatine kinase level. Angina pectoris was defined as coronary insufficiency and was confirmed by exercise stress testing. Thromboembolic, and not hemorrhagic or vasculitic, stroke was confirmed by computed tomography or magnetic resonance imaging. Intermittent claudication was defined as the presence of peripheral atherosclerosis and was confirmed by angiography. The study was approved by the Ethics Committee of Karolinska University Hospital. All subjects gave informed consent before entering the study. Study protocol. The investigation included a written questionnaire, an interview, a physical examination by a rheumatologist, laboratory assessments of blood samples obtained while the patients were fasting, and ultrasound examination of the carotid arteries evaluated by an investigator who was blinded to the subject’s study group. SLE disease activity was determined using the Systemic Lupus Activity Measure (SLAM) (23). Organ damage was determined using the Systemic Lupus International Collaborating Clinics/ACR damage index (24). Insulin sensitivity was estimated by homeostasis model assessment (25). Carotid ultrasound. The right and left carotid arteries were examined with an Acuson Sequoia duplex scanner (Siemens, Mountain View, CA), and the intima-media thickness (IMT) was determined as described elsewhere (26). A plaque was defined as local thickening of the intima media, with an IMT ⬎1 mm. Routine laboratory tests. Plasma lipoprotein concentrations, levels of homocysteine and insulin, insulin resistance, and autoantibodies against DNA, cardiolipin, and ␤2glycoprotein I were determined by routine techniques as described previously (4,27). Determination of IgG and IgM AECA levels. AECAs were detected by enzyme immunoassay as described previously, with some modifications (12). Cryopreserved pooled human umbilical vein endothelial cells (HUVECs) at passage 2 were purchased from Cascade Biologics (Portland, OR). The cultures were maintained at 37°C in an atmosphere of 5% CO2 in endothelial growth medium (EGM phenol red–free medium; Clonetics, San Diego, CA), containing 2% volume/ volume fetal bovine serum and supplements under humidified conditions. HUVECs were seeded on 96-well flat-bottomed tissue culture plates (TPP, Trasadingen, Switzerland) at a density of 1 ⫻ 104 cells/well. After 48 hours, the cells were washed 4 times with phosphate buffered saline (PBS), pH 7.4, and fixed for 15 minutes at room temperature with 0.2% glutaraldehyde in PBS. The fixed cells were washed 4 times with washing buffer containing 0.2% bovine serum albumin (BSA; Sigma, St. Louis, MO) in PBS. Nonspecific binding sites were blocked for 1 hour at room temperature with 1% weight/volume BSA–PBS and 0.1M glycine (Sigma). Serum samples were diluted 1:50 in washing buffer, and 100 l of this dilution was added to each well. After PAF-AH AND OTHER RISK AND PROTECTIVE FACTORS FOR CVD IN SLE 2871 Table 1. Characteristics of the study subjects at baseline* Age, years Disease duration, years SLAM score Body mass index, kg/m2 Waist-to-hip ratio Diabetes mellitus, no. of subjects Nephritis ever, no. of subjects Intima-media thickness, mm Plaque occurrence, no. of subjects Plasma cholesterol, mmoles/liter Plasma triglycerides, mmoles/liter Population controls (n ⫽ 26) SLE cases (n ⫽ 26) SLE controls (n ⫽ 26) 52.3 ⫾ 8.2 – – 24.0 ⫾ 5.0 0.81 ⫾ 0.09 1 0 0.59 ⫾ 0.12 3 5.06 ⫾ 0.93 1.01 ⫾ 0.37 52.2 ⫾ 8.2 20.0 ⫾ 9.9 5 23.8 ⫾ 3.6 0.86 ⫾ 0.08† 3 14 0.66 ⫾ 0.15‡ 17§ 4.99 ⫾ 0.95 1.64 ⫾ 1.00# 52.2 ⫾ 8.2 18.5 ⫾ 9.5 6 24.0 ⫾ 3.6 0.86 ⫾ 0.09 1 9 0.60 ⫾ 0.14 10¶ 5.09 ⫾ 1.14 0.96 ⫾ 0.37 * Disease activity was evaluated by the Systemic Lupus Activity Measure (SLAM). For definition of plaque occurrence, see Patients and Methods. Values are the mean ⫾ SD, except for the SLAM scores, which are the median. Statistical comparisons were made by analysis of variance or chi-square tests. † P ⫽ 0.07 versus population controls. ‡ P ⫽ 0.02 versus population controls and P ⫽ 0.07 versus SLE controls. § P ⫽ 0.001 versus population controls and P ⫽ 0.05 versus SLE controls. ¶ P ⫽ 0.02 versus population controls. # P ⫽ 0.0005 versus population controls and versus SLE controls. overnight incubation at 4°C, alkaline phosphatase–conjugated goat anti-human IgG (1:4,000 dilution) or goat anti-human IgM (1:7,000 dilution) antibodies (both from Sigma) were added. Absorbances at 405 nm were measured by an enzymelinked immunosorbent assay (ELISA) reader. AECA levels (expressed as units) were calculated from a standard curve that had been generated using a pool of positive control sera, the concentration of which had been assigned an arbitrary value of 1,000 units. Measurement of HSP-related factors. Serum levels of Hsp60 and Hsp70 were determined by enzyme immunoassay as described elsewhere (8). Briefly, 96-well microtiter plates were coated with monoclonal antibodies to human Hsp60 (clone LK.1; Stressgen, Victoria, British Columbia, Canada) or Hsp70 (clone C92F3A-5; Stressgen). Plates were washed and blocked with 1% BSA. Samples were added, and bound HSP was detected using rabbit polyclonal anti-Hsp60 or anti-Hsp70 antibody (Stressgen). Bound polyclonal antibody was detected using alkaline phosphatase–conjugated monoclonal antibody to rabbit immunoglobulins (Sigma), followed by p-nitrophenyl phosphate substrate (PNPP; Sigma). The absorbance was measured at 405 nm. Standard dose-response curves were generated using recombinant human Hsp60 or Hsp70 (Stressgen), and the concentrations of Hsp60 and Hsp70 were determined by reference to these standard curves using AssayZap data analysis software (Biosoft, Palo Alto, CA). The interassay variability of the Hsp60 and Hsp70 immunoassays was ⬍10%. HSP antibody levels were determined as described previously (8). Microtiter plates were coated with recombinant human Hsp60 (Stressgen), recombinant human Hsp70 (Stressgen), or recombinant Mycobacterium bovis Hsp65 (kindly provided by Dr. M. Singh, UNDP/World Bank/World Health Organization Special Program for Research and Training in Tropical Diseases, Geneva, Switzerland). Plates were washed and blocked. Samples (typical dilution 1:100) were added, and bound antibodies were detected using alkaline phosphatase– conjugated polyclonal goat anti-human IgA, IgG, and IgM (Sigma) followed by PNPP substrate. Antibody concentrations were determined by comparison with a standard curve that had been generated using samples of predetermined high levels that had been assigned a concentration of 1,000 arbitrary units per milliliter. Measurement of PLA2. Serum levels of PLA2 GIIA were determined using a commercially available kit (human sPLA2 ELISA, catalog no. 1666355; Boehringer, Mannheim, Germany) according to the manufacturer’s instructions. The sensitivity of the assay was ⬍4.4 ng/ml, with a detection range of 1–500 ng/ml. Measurement of PAF-AH. PAF-AH activity was measured using the trichloroacetic acid precipitation procedure, as previously described (28). Assays were performed in 96-well plates. Plasma that had been stored at –80°C was diluted 1:100 in 90 l of PAF-AH assay buffer (pH 7.4), and 10 l of 500 M 3 H-labeled acetyl-PAF (mean ⫾ SD specific activity 81,000 ⫾ 2,000 disintegrations per minute/nmole; DuPont NEN, Boston, MA) was added. Samples were incubated in duplicate for 10 minutes at 37°C, and after precipitation, the radioactivity was assessed in the supernatant. The activity of PAF-AH is expressed in nanomoles of PAF hydrolyzed/minute/ml of plasma. A pool of control plasma (n ⫽ 10) served as an internal standard for all measurements. Measurement of lipids. Levels of blood lipids and ox-LDL were measured as described previously (4); the actual lipid and ox-LDL levels were reported in that article as well. We used these measurements to determine their relationship to PAF-AH. Statistical analysis. Statistical analyses were performed using StatView software (SAS Institute, Gothenburg, Sweden). Skewed continuous variables were logarithmically transformed to attain a normal distribution. Study groups were compared using analysis of variance for continuous variables 2872 CEDERHOLM ET AL Figure 1. Distribution of platelet-activating factor–acetylhydrolase (PAF-AH) activity in population controls (PC), systemic lupus erythematosus (SLE) cases with cardiovascular disease, and SLE controls without cardiovascular disease. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines outside the boxes represent the 10th and the 90th percentiles. Lines inside the boxes represent the 50th percentile. Circles indicate outliers. and the chi-square test for categorical variables. Fisher’s protected least significant difference was used as a post hoc test. Correlation coefficients were calculated using simple regression or distributed variables Spearman’s rank correlation for non-normally distributed data. The significance level was set at P ⬍ 0.05. RESULTS Clinical and metabolic characteristics. Basic clinical and metabolic characteristics of the study subjects are presented in Table 1. As previously reported (4,27), SLE cases had elevated levels of very low-density lipoprotein and Lp(a), decreased levels of high-density lipoprotein (HDL) cholesterol, increased levels of acutephase reactants and erythrocyte sedimentation rates, and elevated levels of lupus anticoagulants and homocysteine as compared with the SLE controls and the population controls. Blood pressure, cumulative (packyears) or present numbers of cigarettes smoked, and prevalence of diabetes mellitus did not differ significantly between the 3 groups (27). SLE cases had taken a higher cumulative dose of prednisolone (P ⫽ 0.05) (4,27,29); however, there was no significant difference in the prednisolone dosage at the time of study compared with that in the SLE controls. Therapy with lipid-lowering agents (statins in all cases), antihypertensive agents, low-dose aspirin, warfarin, and azathioprine was more common among SLE cases (4,29). There was no association between PAF-AH activity or other nontraditional risk factors measured in this study and medication (data not shown). Plasma concentrations of PAF-AH, PLA2, HSPrelated factors, and AECAs. PAF-AH activity in the 3 study groups is shown in Figure 1. Levels of sPLA2, PAF-AH, serum Hsp60 and Hsp70, and anti-Hsp60, anti-Hsp70, anti-Hsp65, and AECAs in the 3 groups are presented in Table 2. PAF-AH activity in SLE cases was higher, although not statistically significantly higher, in the SLE cases than in the SLE controls and the population controls. PAF-AH was negatively associated with Hsp60 (r ⫽ –0.43, P ⫽ 0.037) and Hsp70 (r ⫽ –0.34, P ⫽ 0.099), although the association was not significant for Hsp70. No significant associations between PAF-AH and the other factors we studied were noted. Relationship between PAF-AH activity and clinical and metabolic risk factors for CVD. Correlations between the plasma level of PAF-AH and the levels of lipids, including lipoproteins and ox-LDL, are presented in Table 3. A different pattern of associations was noted between the SLE cases and the 2 control groups. PAF-AH was strongly and positively associated with LDL cholesterol, total cholesterol levels, and interestingly, with ox-LDL (as determined using the monoclonal antibody EO6). However, these associations were not present in the SLE control group; indeed, there was a nonsignificant negative association with all 3 lipids in the SLE control group. A significant negative association with HDL cholesterol was also noted in the SLE control group. There were no associations between PAF-AH and acute-phase reactants, anti–ox-LDL, or IMT (data not shown) in any of the groups studied. DISCUSSION This study is the first to demonstrate that PAF-AH activity in women with SLE and CVD is 30% higher than that in age-matched women with SLE, but without a history of CVD. There was also a trend toward an increased level of PAF-AH in SLE cases compared PAF-AH AND OTHER RISK AND PROTECTIVE FACTORS FOR CVD IN SLE 2873 Table 2. Novel risk and protection factors, by study group* Hsp60, ng/ml Mean ⫾ SD Median Interquartile range Hsp70, ng/ml Mean ⫾ SD Median Interquartile range Anti-Hsp60 antibody, AU/ml Mean ⫾ SD Median Interquartile range Anti-Hsp70 antibody, AU/ml Mean ⫾ SD Median Interquartile range Anti-Hsp65 antibody, AU/ml Mean ⫾ SD Median Interquartile range sPLA2, ng/ml Mean ⫾ SD Median Interquartile range PAF-HA, nmoles/minute/ml of plasma Mean ⫾ SD Median Interquartile range IgG AECA, units Mean ⫾ SD Median Interquartile range IgM AECA, units Mean ⫾ SD Median Interquartile range SLE cases (n ⫽ 26) SLE controls (n ⫽ 26) Population controls (n ⫽ 26) 425.9 ⫾ 311.4 348.6 333.1 500.3 ⫾ 794.7 295.6 214.8 370.03 ⫾ 255.6 330.95 350.2 239.4 ⫾ 367.9 101.9 218.8 115.4 ⫾ 120.1 76.7 84.3 176.82 ⫾ 235.5 78.85 149.2 143.1 ⫾ 146.6 94.3 182.8 92.9 ⫾ 79.4 57.13 85.2 107.07 ⫾ 139.57 60.17 99.3 6.8 ⫾ 3.8 5.4 5.1 7.1 ⫾ 3.9 6.62 3.58 7.13 ⫾ 3.58 7.13 3.47 511.8 ⫾ 255.8 496.7 354.1 494.8 ⫾ 343.5 391.6 377.4 659.51 ⫾ 395.6 530.9 539.5 16.02 ⫾ 3.08 16.6 3.2 15.98 ⫾ 2.4 16.39 2.8 14.9 ⫾ 2.63 14.3 3.75 44.0 ⫾ 16.5 40.4 18.3 34.0 ⫾ 8.8 34.4 12.4 37.6 ⫾ 10.8 36.5 11.1 702.4 ⫾ 285.5 726.0 338.0 652.6 ⫾ 179.8 643.5 221.0 614.8 ⫾ 260.7 582.0 302.2 684.6 ⫾ 314.3 740.5 505.0 635.9 ⫾ 261.7 607.5 352.0 637.9 ⫾ 259.5 609.5 304.0 * Statistical comparisons were made by analysis of variance with Fisher’s post hoc test. Only the platelet-activating factor–acetylhydrolase (PAF-AH) level in the systemic lupus erythematosus (SLE) cases versus the SLE controls was found to be significantly different (P ⫽ 0.005). AU ⫽ arbitrary units; sPLA2 ⫽ secretory phospholipase A2; AECA ⫽ anti–endothelial cell antibody. with controls. Our observations could have several implications. PAF is an important proinflammatory phospho- lipid that is implicated in the etiology of atherosclerosis and CVD. One reason for this is that PAF and PAF-like lipids are involved in the proinflammatory function of Table 3. Relationship between plasma concentrations of PAF-AH and lipids, by study group* Population controls Plasma triglycerides Total cholesterol LDL cholesterol HDL cholesterol Lipoprotein(a) Oxidized LDL SLE cases SLE controls r P r P r P 0.022 0.37 0.083 0.15 0.036 ⫺0.18 0.91 0.06 0.68 0.47 0.81 0.37 0.10 0.54 0.50 0.003 0.38 0.40 0.62 0.0045† 0.0093† 0.99 0.057 0.043† 0.075 ⫺0.23 ⫺0.18 ⫺0.41 ⫺0.05 ⫺0.23 0.71 0.25 0.28 0.038† 0.8 0.26 * Values are Pearson’s product-moment or Spearman’s rank correlation coefficients. PAF-AH ⫽ platelet-activating factor– acetylhydrolase; SLE ⫽ systemic lupus erythematosus; LDL ⫽ low-density lipoprotein; HDL ⫽ high-density lipoprotein. † Statistically significant correlation. 2874 ox-LDL (18,30–32). In addition, PAF is produced by activated platelets, monocytes, and endothelial cells and could play an important role in endothelial activation (30). The potential involvement of PAF in atherogenesis has been suggested by the observation that inhibiting PAF activity decreases the development of atherosclerosis in animal models (33). PAF-AH is produced by cells of monocyte/ macrophage origin, as well as by T cells and mast cells, all of which are present in atherosclerotic lesions (34). PAF-AH degrades PAF and PAF-like lipids by hydrolyzing its acetate moiety in the sn-2 position of the phospholipids, and its capacity to degrade PAF prompted the proposition that PAF-AH was antiatherogenic. Indeed, animal models have shown that PAF-AH is antiinflammatory and antiatherogenic (35). However, PAF-AH cannot bind to LDL in mice because of differences in amino acid composition in the 114–117 domain (36). Data from mouse models of atherosclerosis are therefore not truly reflective of the clinical situation in this context, since in humans, 70% of PAF-AH binds to LDL cholesterol, with the remaining PAF-AH being transported by HDL cholesterol (37). Instead, an increasing body of evidence indicates that PAF-AH may be either a risk factor or a marker for CVD, including stroke and coronary artery disease (38–40). Our data support this proposition. An interesting observation from the present study is that PAF-AH activity in SLE cases was not only associated with total cholesterol and LDL cholesterol, but also with ox-LDL (as determined on the basis of EO6 epitope concentration). It has been suggested that PAF-AH is atherogenic, since it is the only lipoproteinassociated enzyme that can bind to LDL cholesterol and, as a consequence, become trapped in the arterial wall, within which it can induce the hydrolysis of oxidized phospholipids. However, LPC and other oxidized phospholipid derivatives produced by PAF-AH may also be highly atherogenic. LPC promotes not only the chemotaxis of monocytes (41), but also the activation of T cells and their production of proinflammatory cytokines (15), which are abundant in lesions (1). Our finding is therefore compatible with the possibility that PAF-AH promotes atherogenesis by increasing the concentration of LPC in the arterial wall. In sharp contrast to the situation in SLE cases, there was no statistically significant relationship between PAF-AH and total cholesterol, LDL cholesterol, or ox-LDL in the SLE controls. Indeed, although not of statistical significance, the relationship between PAF-AH and total cholesterol, LDL cholesterol, or CEDERHOLM ET AL ox-LDL appeared to be a negative one. This might indicate that PAF-AH plays a different role in the 2 groups, and it is possible that this differential role relates to different patterns of association with lipoproteins. However, the basis for the negative association between PAF-AH and HDL cholesterol in SLE controls is unclear. Consistent with the findings of previous investigators (38–40), we found that PAF-AH was not associated with common markers of inflammation in any of the groups studied. Circulating PLA2 enzymes, including sPLA2, have been described in patients with atherosclerosis and CVD. The sPLA2 enzyme promotes the modification of phospholipids, including those present in circulating lipoproteins. In contrast to PAF-AH, which acts only on PAF and PAF-like lipids, sPLA2 also hydrolyzes intact phospholipids, thereby generating lysophospholipids and free fatty acids (42). PLA2 is present in atherosclerotic lesions (14,42,43), and may promote atherosclerosis by acting on LDL cholesterol trapped in the vascular wall via binding to proteoglycans and inducing oxidation of LDL (42). PLA2 may also be atherogenic and increase the risk of CVD by inducing the formation of small, dense LDL particles in the circulation (42). Although PLA2 levels were not increased in the circulation of the SLE cases in our study, this negative finding does not necessarily rule out the possibility that sPLA2 plays a role in artery wall damage in SLE. The role of HSP-related factors in CVD in general is likely to be complex. The proposition that anti-Hsp65 is atherogenic is supported by the results of animal experiments and clinical studies. One explanation for this finding could be that HSPs are evolutionarily conserved molecules and that anti-HSPs could cross-react with HSPs present both on activated endothelial cells and in bacteria, including mycobacteria (6,10,44). Recently, HSPs, including Hsp60 and Hsp70, have been identified in the circulation, and we have reported that Hsp60 is significantly increased in early CVD occurring in patients with borderline hypertension (8). We have also demonstrated that Hsp70 appears to protect against atherosclerosis, since high levels of Hsp70 are associated with a decreased progression of atherosclerosis in patients with established hypertension (9). However, in this cross-sectional study of SLErelated CVD, HSP-related factors were not increased. Prospective studies are necessary to elucidate the exact role of these factors in SLE-related CVD, and it is still possible that a defective HSP response could promote atherothrombosis in SLE. PAF-AH AND OTHER RISK AND PROTECTIVE FACTORS FOR CVD IN SLE Antibodies against endothelial cells are clearly implicated in SLE and are associated with disease activity and vasculitis (45). Furthermore, they promote endothelial activation by acting directly on endothelial cells (46). AECAs have also been implicated in CVD in the general population (12,47), and levels are increased in SLE patients. These antibodies also cross-react with a pivotal antigen in atherosclerosis, ox-LDL (20). However, we found that AECAs were not significantly increased in patients with SLE-related CVD in our study. The capacity of AECAs to promote atherogenesis in patients with more active SLE remains to be evaluated. In conclusion, PAF-AH may play an important role in SLE-related CVD by promoting a proinflammatory state and LDL modification of the artery wall. It remains puzzling, however, why the generation of LPC and oxidized short fatty acids would be more atherogenic than intact PAF and/or oxidized phospholipids. By analogy to the increase in specific antibodies during infection, one may also envision that the balance between these substances may be of importance (48) or that the increase in PAF-AH is protective. Other nontraditional immune-related factors tested in this study appear to play a lesser role in SLE-related CVD than in CVD in the general population. Further prospective studies are necessary to clarify whether these and other factors are responsible for the elevated risk of CVD in patients with SLE. ACKNOWLEDGMENTS We are grateful to Mikael Heimbürger for referring patients from Huddinge University Hospital, to Jill Gustafsson and Eva Jemseby for their help with management of the patient cohorts and the blood sampling, to Angela Silveira and Anders Hamsten for lipoprotein and insulin determinations, to Joseph Witztum for help with ox-LDL determinations, and to Kerstin Jensen-Urstad for ultrasound measurements. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. REFERENCES 1. Frostegard J, Ulfgren AK, Nyberg P, Hedin U, Swedenborg J, Andersson U, et al. Cytokine expression in advanced human atherosclerotic plaques: dominance of pro-inflammatory (Th1) and macrophage-stimulating cytokines. Atherosclerosis 1999;145: 33–43. 2. Urowitz MB, Bookman AA, Koehler BE, Gordon DA, Smythe HA, Ogryzlo MA. The bimodal mortality pattern of systemic lupus erythematosus. Am J Med 1976;60:221–5. 3. Manzi S, Meilahn EN, Rairie JE, Conte CG, Medsger TA Jr, Jansen-McWilliams L, et al. Age-specific incidence rates of myocardial infarction and angina in women with systemic lupus erythematosus: comparison with the Framingham Study. Am J Epidemiol 1997;145:408–15. 4. Svenungsson E, Jensen-Urstad K, Heimburger M, Silveira A, 20. 21. 22. 2875 Hamsten A, de Faire U, et al. Risk factors for cardiovascular disease in systemic lupus erythematosus. Circulation 2001;104: 1887–93. Cullen P, Baetta R, Bellosta S, Bernini F, Chinetti G, Cignarella A, et al. Rupture of the atherosclerotic plaque: does a good animal model exist. Arterioscler Thromb Vasc Biol 2003;23:535–42. Frostegard J, Lemne C, Andersson B, van der Zee R, Kiessling R, de Faire U. Association of serum antibodies to heat-shock protein 65 with borderline hypertension. Hypertension 1997;29:40–4. Frostegard J, Kjellman B, Gidlund M, Andersson B, Jindal S, Kiessling R. Induction of heat shock protein in monocytic cells by oxidized low density lipoprotein. Atherosclerosis 1996;121:93–103. Pockley AG, Wu R, Lemne C, Kiessling R, de Faire U, Frostegard J. Circulating heat shock protein 60 is associated with early cardiovascular disease. Hypertension 2000;36:303–7. Pockley AG, Georgiades A, Thulin T, de Faire U, Frostegard J. Serum heat shock protein 70 levels predict the development of atherosclerosis in subjects with established hypertension. Hypertension 2003;42:235–8. Xu Q, Willeit J, Marosi M, Kleindienst R, Oberhollenzer F, Kiechl S, et al. Association of serum antibodies to heat-shock protein 65 with carotid atherosclerosis. Lancet 1993;341:255–9. Wick G, Perschinka H, Millonig G. Atherosclerosis as an autoimmune disease: an update. Trends Immunol 2001;22:665–9. Frostegard J, Wu R, Gillis-Haegerstrand C, Lemne C, de Faire U. Antibodies to endothelial cells in borderline hypertension. Circulation 1998;98:1092–8. Goudevenos J, Tselepis AD, Vini MP, Michalis L, Tsoukatos DC, Elisaf M, et al. Platelet-associated and secreted PAF-acetylhydrolase activity in patients with stable angina: sequential changes of the enzyme activity after angioplasty. Eur J Clin Invest 2001;31: 15–23. Elinder LS, Dumitrescu A, Larsson P, Hedin U, Frostegard J, Claesson HE. Expression of phospholipase A2 isoforms in human normal and atherosclerotic arterial wall. Arterioscler Thromb Vasc Biol 1997;17:2257–63. Huang YH, Schafer-Elinder L, Wu R, Claesson HE, Frostegard J. Lysophosphatidylcholine (LPC) induces proinflammatory cytokines by a platelet-activating factor (PAF) receptor-dependent mechanism. Clin Exp Immunol 1999;116:326–31. Xu Q, Schett G, Seitz CS, Hu Y, Gupta RS, Wick G. Surface staining and cytotoxic activity of heat-shock protein 60 antibody in stressed aortic endothelial cells. Circ Res 1994;75:1078–85. Binder CJ, Chang MK, Shaw PX, Miller YI, Hartvigsen K, Dewan A, et al. Innate and acquired immunity in atherogenesis. Nat Med 2002;8:1218–26. Frostegard J, Huang YH, Ronnelid J, Schafer-Elinder L. Plateletactivating factor and oxidized LDL induce immune activation by a common mechanism. Arterioscler Thromb Vasc Biol 1997;17: 963–8. Watson AD, Navab M, Hama SY, Sevanian A, Prescott SM, Stafforini DM, et al. Effect of platelet activating factor-acetylhydrolase on the formation and action of minimally oxidized low density lipoprotein. J Clin Invest 1995;95:774–82. Wu R, Svenungsson E, Gunnarsson I, Haegerstrand-Gillis C, Andersson B, Lundberg I, et al. Antibodies to adult human endothelial cells cross-react with oxidized low-density lipoprotein and ␤2-glycoprotein I (␤2-GPI) in systemic lupus erythematosus. Clin Exp Immunol 1999;115:561–6. Praprotnik S, Blank M, Meroni PL, Rozman B, Eldor A, Shoenfeld Y. Classification of anti–endothelial cell antibodies into antibodies against microvascular and macrovascular endothelial cells: the pathogenic and diagnostic implications. Arthritis Rheum 2001;44:1484–94. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7. 2876 23. Liang MH, Socher SA, Roberts WN, Esdaile JM. Measurement of systemic lupus erythematosus activity in clinical research. Arthritis Rheum 1988;31:817–25. 24. Gladman D, Ginzler E, Goldsmith C, Fortin P, Liang M, Urowitz M, et al. The development and initial validation of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index for systemic lupus erythematosus. Arthritis Rheum 1996;39:363–9. 25. Matthews DR, Hosker JP, Rudenski AS, Naylor BR, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance beta-cell function from fasting glucose and insulin concentrations in man. Diabetologia 1985;28:412–9. 26. Lemne C, Jogestrand T, de Faire U. Carotid intima-media thickness and plaque in borderline hypertension. Stroke 1995;26:34–9. 27. Svenungsson E, Fei GZ, Jensen-Urstad K, de Faire U, Hamsten A, Frostegard J. TNF-␣: a link between hypertriglyceridaemia and inflammation in SLE patients with cardiovascular disease. Lupus 2003;12:454–61. 28. Tselepis AD, Karabina SA, Stengel D, Piedagnel R, Chapman MJ, Ninio E. N-linked glycosylation of macrophage-derived PAF-AH is a major determinant of enzyme association with plasma HDL. J Lipid Res 2001;42:1645–54. 29. Svenungsson E, Gunnarsson I, Fei GZ, Lundberg IE, Klareskog L, Frostegard J. Elevated triglycerides and low levels of high-density lipoprotein as markers of disease activity in association with up-regulation of the tumor necrosis factor ␣/tumor necrosis factor receptor system in systemic lupus erythematosus. Arthritis Rheum 2003;48:2533–40. 30. Prescott SM, Zimmerman GA, Stafforini DM, McIntyre TM. Platelet-activating factor and related lipid mediators. Annu Rev Biochem 2000;69:419–45. 31. Watson AD, Leitinger N, Navab M, Faull KF, Horkko S, Witztum JL, et al. Structural identification by mass spectrometry of oxidized phospholipids in minimally oxidized low density lipoprotein that induce monocyte/endothelial interactions and evidence for their presence in vivo. J Biol Chem 1997;272:13597–607. 32. Marathe GK, Davies SS, Harrison KA, Silva AR, Murphy RC, Castro-Faria-Neto H, et al. Inflammatory platelet-activating factor-like phospholipids in oxidized low density lipoproteins are fragmented alkyl phosphatidylcholines. J Biol Chem 1999;274: 28395–404. 33. Subbanagounder G, Leitinger N, Shih PT, Faull KF, Berliner JA. Evidence that phospholipid oxidation products and/or plateletactivating factor play an important role in early atherogenesis: in vitro and In vivo inhibition by WEB 2086. Circ Res 1999;85:311–8. 34. Caslake MJ, Packard CJ. Lipoprotein-associated phospholipase A2 (platelet-activating factor acetylhydrolase) and cardiovascular disease. Curr Opin Lipidol 2003;14:347–52. 35. Quarck R, De Geest B, Stengel D, Mertens A, Lox M, Theilmeier G, et al. Adenovirus-mediated gene transfer of human plateletactivating factor-acetylhydrolase prevents injury-induced neointima formation and reduces spontaneous atherosclerosis in apolipoprotein E-deficient mice. Circulation 2001;103:2495–500. 36. Stafforini DM, Tjoelker LW, McCormick SP, Vaitkus D, McIntyre CEDERHOLM ET AL 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. TM, Gray PW, et al. Molecular basis of the interaction between plasma platelet-activating factor acetylhydrolase and low density lipoprotein. J Biol Chem 1999;274:7018–24. Tselepis AD, Dentan C, Karabina SA, Chapman MJ, Ninio E. PAF-degrading acetylhydrolase is preferentially associated with dense LDL and VHDL-1 in human plasma: catalytic characteristics and relation to the monocyte-derived enzyme. Arterioscler Thromb Vasc Biol 1995;15:1764–73. Satoh K, Yoshida H, Imaizumi T, Takamatsu S, Mizuno S. Platelet-activating factor acetylhydrolase in plasma lipoproteins from patients with ischemic stroke. Stroke 1992;23:1090–2. Blankenberg S, Stengel D, Rupprecht HJ, Bickel C, Meyer J, Cambien F, et al. Plasma PAF-acetylhydrolase in patients with coronary artery disease: results of a cross-sectional analysis. J Lipid Res 2003;44:1381–6. Packard CJ, O’Reilly DS, Caslake MJ, McMahon AD, Ford I, Cooney J, et al, for the West of Scotland Coronary Prevention Study Group. Lipoprotein-associated phospholipase A2 as an independent predictor of coronary heart disease. N Engl J Med 2000;343:1148–55. Quinn MT, Parthasarathy S, Steinberg D. Lysophosphatidylcholine: a chemotactic factor for human monocytes and its potential role in atherogenesis. Proc Natl Acad Sci U S A 1988;85: 2805–9. Hurt-Camejo E, Camejo G, Peilot H, Oorni K, Kovanen P. Phospholipase A2 in vascular disease. Circ Res 2001;89:298–304. Hurt-Camejo E, Andersen S, Standal R, Rosengren B, Sartipy P, Stadberg E, et al. Localization of nonpancreatic secretory phospholipase A2 in normal and atherosclerotic arteries: activity of the isolated enzyme on low-density lipoproteins. Arterioscler Thromb Vasc Biol 1997;17:300–9. Xu Q, Dietrich H, Steiner HJ, Gown AM, Schoel B, Mikuz G, et al. Induction of arteriosclerosis in normocholesterolemic rabbits by immunization with heat shock protein 65. Arterioscler Thromb 1992;12:789–99. D’Cruz DP, Houssiau FA, Ramirez G, Baguley E, McCutcheon J, Vianna J, et al. Antibodies to endothelial cells in systemic lupus erythematosus: a potential marker for nephritis and vasculitis. Clin Exp Immunol 1991;85:254–61. Carvalho D, Savage CO, Isenberg D, Pearson JD. IgG anti–endothelial cell autoantibodies from patients with systemic lupus erythematosus or systemic vasculitis stimulate the release of two endothelial cell–derived mediators, which enhance adhesion molecule expression and leukocyte adhesion in an autocrine manner. Arthritis Rheum 1999;42:631–40. D’Anastasio C, Impallomeni M, McPherson GA, Clements WG, Howells GL, Brooks PA, et al. Antibodies against monocytes and endothelial cells in the sera of patients with atherosclerotic peripheral arterial disease. Atherosclerosis 1988;74:99–105. Tsoukatos DC, Liapikos TA, Tselepis AD, Chapman MJ, Ninio E. Platelet-activating factor acetylhydrolase and transacetylase activities in human plasma low-density lipoprotein. Biochem J 2001; 357:457–64.